KR102335738B1 - Method and apparatus for calculating a rotating angle of a device - Google Patents

Abstract

A device for determining a rotation angle with respect to a moving direction, the device comprising: an acceleration sensing unit configured to obtain acceleration information on the movement of the device; a storage unit for storing the obtained acceleration information; and a control unit, wherein the control unit transforms the coordinates of the obtained acceleration information, filters the acceleration information on the transformed coordinates, and advances the device using the gravitational axis acceleration information on the transformed coordinates. A rotation angle with respect to a direction is determined, and the gravitational acceleration information may be derived using information about a time point at which a vertical acceleration component on the coordinates of the device converted from the acceleration information is maximum.

Description

Method and apparatus for determining a rotation angle of a device {Method and apparatus for calculating a rotating angle of a device}

The present invention relates to a method for determining a rotation angle of a device and an apparatus therefor. More particularly, it relates to a method of determining a misaligned angle of a device with respect to a traveling direction of the device.

A Pedestrian Dead Reckoning (PDR) method may be used as a location estimation method using a device such as a smartphone. In the location estimation performance of pedestrian navigation, the rotation angle (misaligned angle) of the device is a variable that has a large influence, and even if the moving distance is accurately calculated when the misaligned angle of the device is measured incorrectly, a large error may occur depending on the location. .

In particular, when a user of a small device, such as a smartphone, grips and moves the device, a misaligned angle with respect to the moving direction of the device may be arbitrarily changed according to the user's convenience or intention. For example, when a smartphone user grips and moves the long axis of the smartphone in a portrait mode or a landscape mode, and about 60 degrees from the direction of travel, the smart phone in the yaw direction When the phone moves in a distorted state, it is difficult to measure the skew angle therebetween using only the acceleration sensor mounted on the smartphone. Due to this inaccurate angle measurement, a large error may also occur in the position estimation.

For example, when the user walks north with the device, if the heading of the device is to the north, the user's moving trajectory is estimated to be north. However, in the same situation, if the heading of the device faces east, the user's movement trajectory is estimated to the east even though the user walks toward the north, and a large position error may occur. Therefore, in this case, the angle between the user's heading and the device's heading should be estimated and compensated for.

The reason that it is difficult to measure the misaligned angle of the device with respect to the moving direction of the device is that the accelerometer mounted on the device senses both the acceleration information in the direction of the user's movement and the acceleration information in the direction of the user's movement. because it cannot be distinguished.

Conventionally, acceleration information sensed using an acceleration sensor, a gyroscope, and a geomagnetic sensor is interpreted in the frequency domain, or a misaligned angle of the device with respect to the user's moving direction is measured using the user's movement dynamic characteristics. did

The method of measuring the misaligned angle of the device using the reaction acceleration vector at the moment when the user steps forward to move is difficult to distinguish the reaction acceleration vector due to sensor noise. There is a case where a large angle error occurs. The error can be reduced by filtering the final measurement angle information using a low pass filter (LPF), but on the other hand, there may be a problem in that the measurement time is delayed due to the filtering.

The present disclosure proposes a method of efficiently classifying gravitational acceleration from sensing information of only an acceleration sensor by identifying a movement characteristic when a user moves. In the case of measurement according to the proposed method, there is an advantageous effect of being able to more accurately calculate the misaligned angle of the device with respect to the user's moving direction compared to the existing methods by classifying the gravitational acceleration component from the acceleration information. This is a technology to perform location estimation with low power by increasing the location estimation accuracy of the mobile phone and minimizing sensing power.

The present invention relates to a method for determining an angle between a moving direction of a user and a heading direction of a device, and the rotation angle of the device can be more accurately measured using an acceleration component.

In accordance with an embodiment of the present invention, there is provided a method for determining a rotation angle with respect to a moving direction of a device, the method comprising: obtaining acceleration information on the movement of the device from an acceleration sensor of the device; transforming the coordinates of the acquired acceleration information of the device; filtering the acceleration information on the transformed coordinates of the device; and determining a rotation angle with respect to the traveling direction of the device by using the information on the acceleration of the axis of gravity on the converted coordinates, wherein the information on the acceleration of the axis of gravity is obtained from the information on the converted coordinates of the device in the acceleration information. It can be derived using information about a time point at which the vertical acceleration component is maximum.

Preferably, a point in time at which the vertical acceleration component on the transformed coordinates is maximum may be the same as a point in time when the magnitude of the obtained acceleration has a maximum value.

Preferably, the minimum value of the vertical acceleration component on the transformed coordinates may be derived using information about a time point having a maximum value of the vertical acceleration component on the transformed coordinates.

Preferably, the horizontal acceleration component included in the acceleration information may be acceleration information for a moving direction of the device, and the vertical acceleration component may be an acceleration component between the device and a horizontal plane.

Preferably, the converting of the obtained coordinates of the acceleration information may include converting the acceleration information from coordinates on a coordinate system based on a device to coordinates on a coordinate system based on a gravity direction.

More preferably, the coordinate system based on the gravitational direction may include three axes orthogonal to each other, and a first axis of the three axes may be an axis parallel to the direction of gravitational acceleration.

More preferably, the second axis of the three axes is defined by rotating one axis on the device reference coordinate system by a roll angle, and the third axis of the three axes pitches another axis on the device reference coordinate system (pitch) can be defined by angular rotation.

Preferably, the vertical acceleration component and the horizontal acceleration component on the transformed coordinates may change with the same period value.

Preferably, the vertical acceleration component and the horizontal acceleration component on the transformed coordinates may have a phase difference of 90 degrees from each other.

More preferably, the vertical acceleration component may be 90 degrees faster than the horizontal acceleration component.

More preferably, the vertical acceleration component may be 90 degrees slower than the horizontal acceleration component.

Preferably, the gravitational axis acceleration information may be derived by averaging the magnitudes of vertical and horizontal acceleration components on the transformed coordinates of the device over a plurality of periods.

Preferably, the filtering of the acceleration information may further include converting the acceleration information into a frequency domain.

More preferably, the filtering of the acceleration information may include bandpass filtering the acceleration information.

More preferably, the bandpass filtering may include filtering based on a center frequency determined using a dynamic characteristic of the device from the acceleration information converted into the frequency domain.

More preferably, the dynamic characteristics of the device may be obtained by a posture in which the device user steps in the moving direction.

In accordance with another embodiment of the present invention, there is provided a device for determining a rotation angle with respect to a moving direction, comprising: an acceleration sensing unit configured to obtain acceleration information on the movement of the device; a storage unit for storing the obtained acceleration information; and a control unit, wherein the control unit transforms the coordinates of the obtained acceleration information, filters the acceleration information on the transformed coordinates, and advances the device using the gravitational axis acceleration information on the transformed coordinates. A rotation angle with respect to a direction is determined, and the gravitational acceleration information may be derived using information about a time point at which a vertical acceleration component on the coordinates of the device converted from the acceleration information is maximum.

1 shows a rotation axis of a device according to an embodiment of the present invention.
2 to 4 show a misaligned state with respect to the traveling direction of the device according to an embodiment of the present invention.
5 shows coordinate systems applicable to an embodiment of the present invention.
6 illustrates a coordinate transformation according to an embodiment of the present invention.
7 illustrates a process of measuring an angle of a device according to an embodiment of the present invention.
8 illustrates a change in vertical axis height according to a user's moving direction according to an embodiment of the present invention.
9 illustrates a change in vertical axis speed according to a user's moving direction according to an embodiment of the present invention.
10 illustrates a change in vertical acceleration according to a user's moving direction according to an embodiment of the present invention.
11 illustrates a change in horizontal acceleration according to a user's moving direction according to an embodiment of the present invention.
12 illustrates a case in which a user moves at a constant speed and a case in which a user moves at a reduced speed according to an embodiment of the present invention.
13 illustrates a coordinate system before and after coordinate transformation according to an embodiment of the present invention.
14 is a graph experimentally showing the Z-axis acceleration in the coordinate system after transformation according to an embodiment of the present invention.
15 is a graph showing that the time point at which the minimum value of the vertical acceleration appears in the local coordinate system may be different from the time point at which the minimum value of the magnitude of the acceleration appears in the simulation.
16 illustrates a method of filtering an acceleration component according to a user's moving direction according to an embodiment of the present invention.
17 is a graph experimentally showing a value of gravitational acceleration in a coordinate system before coordinate transformation according to an embodiment of the present invention.
18 illustrates a method of removing noise through filtering according to an embodiment of the present invention.
19 shows a misaligned angle of a device according to an embodiment of the present invention.
20 is a graph experimentally showing an error value of each misaligned measurement of a device according to an embodiment of the present invention.
21 is a block diagram conceptually illustrating the configuration of a device according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same reference numerals refer to the same components, and the size or thickness of each component may be exaggerated for clarity of description.

In the present disclosure, a “rotating angle” of a device refers to an angle based on a specific coordinate system with respect to a posture positioned in a three-dimensional space. It is used in the same meaning as “a misaligned angle” in a three-dimensional space and may be used interchangeably in the present disclosure. In addition, the “rotation angle” may be expressed as an angle between the axis and an axis on an arbitrarily determined coordinate system.

In the present disclosure, the term “body frame” refers to coordinates where each axis composed of the x, y, and z axes with respect to the device is orthogonal to each other. The long-axis direction of the device is defined as the x-axis, the short-axis direction is defined as the y-axis, and the direction defined by the right-hand rule by the defined x-axis and y-axis is defined as the z-axis. When the four fingers of the right hand are wound from the x-axis to the y-axis, the direction the thumb points is the z-axis. That is, it is a coordinate system in which the top-down direction of a device such as a smartphone is defined as the z-axis.

In the present disclosure, "local frame" is a coordinate system that refers to a specific three-dimensional coordinate system. The characteristics of the local coordinate system will be described in relation to the body coordinate system as follows. First, when rotating by -1 * roll angle around the x-axis of the body frame, the y-axis of the body frame is positioned as the y'-axis in the plane perpendicular to the direction of gravitational acceleration. When the y' axis is rotated by -1 * pitch angle, the x axis is positioned as the x' axis in the plane perpendicular to the direction of gravitational acceleration. The determined x' and y' axes are defined as the x and y axes of the local coordinate system, respectively, and the coordinate system defining the three axes with the direction defined by the right-hand rule by the x and y axes as the Z axis becomes the local coordinate system.

Among the x, y, and z axes of the local coordinate system, the z axis indicates a direction parallel to the gravitational acceleration, and the x axis is an axis indicating 0 degrees when it coincides with the direction of the device. The y-axis is an axis determined by the determined x and x-axis, and the three axes have characteristics that are orthogonal to each other. Therefore, it is preferable to broadly interpret the local coordinate system as a coordinate system including the features described in the present disclosure among coordinate systems having three axes orthogonal to each other in a three-dimensional space.

Hereinafter, a method of measuring a rotation angle of a device according to an embodiment of the present invention will be described in detail with reference to FIGS. 1 to 20 .

1 shows a rotation axis of a device according to an embodiment of the present invention.

Since the device moves in a three-dimensional space, it may be positioned in various directions and angles in space. Since all devices have a physical center, three axes can exist based on this center. Here, the physical center may be defined in various ways, such as a center of gravity, a geometrically defined center, and a location of an acceleration sensor. Three axes exist in the front-back direction, left-right direction, and up-down direction based on the center of the device. These three axes are arbitrarily determined and orthogonal axes, and may be set differently depending on the physical structure of the device. For example, when the device has a rectangular parallelepiped structure such as a smartphone, three axes may be set with the display unit on which the screen is displayed as the front. On the other hand, if the front surface cannot be specified as a sphere-like device, it may be irrelevant no matter what direction is set as a reference.

As shown in FIG. 1 , as a method of defining three axes of a device such as a smartphone, in a portrait direction that faces the display part of the device to display the display part vertically, x based on the center of the device Three axes can be defined: , y, and z.

As described above, each axis composed of the x, y, and z axes with respect to the device is orthogonal to each other and may be called body coordinates or a body frame. The x, y, and z axes of the body coordinate system are fixed to a device such as a smartphone and change together when the attitude of the smartphone in 3D space changes. In the present disclosure, the subscripts of XB, YB, and ZB indicating the x, y and z axes on the notation indicate the first letter “B” of the body frame.

2 to 4 show a misaligned angle with respect to the traveling direction of the device according to an embodiment of the present invention.

As transportation facilities such as automobiles and subways develop, people move more frequently, and people bring devices such as smart phones and tablet PCs with them when they move. Device users can use various applications performed on the device while moving and use services. As a location-based service (LBS) is developed, a user can immediately check a service available in a current location while moving, and can select and use a desired service. Alternatively, users may play content such as games and movies in the device regardless of their current location.

When a user uses a navigation application, information displayed on the device may vary according to a moving direction and a moving speed of the device. Alternatively, when a user uses a game application, a control method and a display method may vary depending on whether the user manipulates the device in a portrait or a landscape mode. Accordingly, there is a need for a method for accurately measuring a misaligned angle with respect to a device. Here, the misaligned angle with respect to the device means the angle between the two vectors when the user's moving direction vector and the device's heading vector are interpreted in a plane perpendicular to the gravity vector.

As shown in FIG. 2 , the user may hold and move the device in a state in which the long axis of the device coincides in the same direction as the moving direction. The user can move while looking at the screen of the device in a portrait (direction in which the long axis direction of the device becomes a heading) and holding it.

For example, when the user executes a navigation application of the smartphone device and moves while being guided, the heading vector of the device may be matched in the same direction as that of the user (or device).

As shown in FIG. 3 , the user can match the heading vector with the device in a direction perpendicular to the direction of the user (or device), and use the device as a landscape (direction in which the short-axis direction of the device becomes the heading direction) method You can also move while looking at the screen. In this case, it can be seen that the rotation state of the device with respect to the moving direction of the user (or device) is different from that of FIG. 2 described above.

As shown in FIG. 4 , the user may move while maintaining an arbitrarily twisted angle without matching the moving direction of the user (or device) with the axis of the device. For example, a user can walk while using a navigation service using a voice signal from a navigation application while putting the device in a bag or trouser pocket. In this case, the heading vector of the device in the bag or trouser pocket may not be fixed so as to match the traveling direction.

5 shows coordinate systems applicable to an embodiment of the present invention.

In order to measure the misaligned angle of the device, the device receives information from various sensors, and processes the received information to obtain desired angle information. In this case, angle information may be expressed on various coordinate systems, and a criterion for measuring an angle may vary depending on which coordinate system is expressed. Accordingly, it is necessary to process information expressed on various coordinate systems to be compatible with each other.

As an example of the coordinate system of Pedestrian Dead Reckoning (PDR), a body frame or a local frame may be described. Note that various coordinate systems exist and are not limited to the two coordinate systems described in the present disclosure, and the two coordinate systems are described as an example for convenience of description.

5( a ) shows a body frame having an arbitrary posture to define a local frame in an embodiment of the present invention. Figure 5 (a), when rotated Coordinates of Y _L axis of the body coordinate system of X _B central axis in the body coordinate system of the Y _B-axis and Z _B axes, the Y _B axis of the body coordinate system is parallel to a plane perpendicular to the direction of gravity in It can be defined as a vector. That is, the Y and _L axis of the local coordinate system may be perpendicular to the gravitational acceleration vector. In this case, there are two cases of the rotation method, and the rotation is rotated in _{a direction in which the angle between the Z B} axis of the body coordinate system and the gravitational acceleration vector does not exceed 90 degrees. This can be illustrated as in Fig. 5(b). When rotated X _L-axis of the local coordinate system, Fig. 5 (b) Coordinates of Y _L axis center in the body coordinate system of X _B-axis and Z _B axes of the X _B axis of the body coordinate system is parallel to a plane perpendicular to the direction of gravity It can be defined as a vector. In this case, there may be two cases of the rotation method, and the Z _B axis of the rotationally transformed body coordinate system may be rotated to coincide with the direction of gravitational acceleration. _{Therefore, the Z B} axis of the local coordinate system is naturally defined in the direction of gravitational acceleration. This can be illustrated as in FIG. 5(c).

The origin of the local coordinate system is the same as the origin defined in the body coordinate system, and the _{subscripts of X L} , Y _{L and Z L in} the local coordinate system can indicate the first letter “L” of the local frame. Accordingly, in the local coordinate system, since the Z _L axis is parallel to the gravitational acceleration axis _{as shown in FIG. 5( c ), only the Z L} axis can sense the gravitational acceleration. The three axes of the local coordinate system are also orthogonal to each other, and from the misaligned posture information between the local coordinate system and the body coordinate system in the three-dimensional coordinate system, the direction and angle of the device with respect to the user's moving direction on the horizontal plane perpendicular to the acceleration of gravity can be expressed.

In order to know the posture in the local coordinate system and the body coordinate system, roll, pitch, and yaw angles should be defined based on the local coordinate system. First, if the roll is described, the rotational motion that occurs around the axis (x-axis) in the front-rear direction according to the linear motion along the axis (y-axis) in the left and right direction of the coordinate system is called rolling. Describing the pitch, the rotational motion occurring around the left-right axis (y-axis) caused by the linear motion along the x-axis of the coordinate system is called pitching. Describing the yaw, a rotational motion about the axis (z-axis) in the vertical direction of the coordinate system is called yaw.

In Fig. 5(d), when the origins of the two coordinate systems coincide, the yaw angle is _{positive when the Z L} axis is wrapped with the right hand according to the right hand rule, and the direction from the palm to the finger is the positive direction, and the sign of the yaw angle is determined and the amount of rotation can directly determine the size of the yaw angle. Here, the yaw angle can always be defined as 0 because there is no Z-axis rotation transformation from the body coordinate system in the local coordinate system. The pitch angle and roll angle are determined by the right-hand rule, which is the same method as the yaw angle with respect to the _{Y L} and X _{L axes, respectively.} In order to transform from the local coordinate system to the body coordinate system, it is important to define the transformation order, and in the present invention, it is defined to transform in the order of yaw, pitch, and roll based on the local coordinate system. The reason is that when converting from the body coordinate system to the local coordinate system, it is defined in the order of -1 * roll angle, -1 * pitch angle, 0 * (-1 * yaw angle). The order of these transformations may vary depending on the definition of the coordinate transformation, only the definition is different, and the result of the transformation of the body coordinate system and the local coordinate system (that is, the posture in each coordinate system) does not change.

It should be noted that the body coordinate system and the local coordinate system are only a matter of choice of which direction is to be determined as a positive direction in the same space. In a coordinate system to which all laws of physics are equally applied, since they are all equal coordinate systems, it is possible to transform between coordinate systems. Also, in the coordinate transformation method, the transformation procedure and transformation formula may change depending on how the coordinate system is defined, how roll, pitch, and yaw angle are defined, and how the coordinate transformation order is defined, but only the expression is different and the meaning of the coordinate transformation is the same. .

6 illustrates a coordinate transformation according to an embodiment of the present invention. In FIG. 6, a method of performing coordinate transformation in consideration of only the gravitational acceleration for a state in which there is no movement of the device will be described.

As shown in FIG. 6 , the acceleration component expressed on the body coordinate system can also be expressed in the local coordinate system. Similarly, information expressed on the local coordinate system can be transformed and expressed on the body coordinate system. Transformation between coordinate systems may vary depending on the value of the gravitational acceleration sensed in the body coordinate system.

In the body coordinate system, in general _{, the gravitational acceleration G B} is sensed by orthogonal projection for each of the three axes of the _{X B} axis, the Y _B axis and the Z _B axis, where the subscript B stands for the body frame. . Thus, the acceleration of gravity G _B may be divided into three components of gx, gy and gz with respect to three axes. That is, the gravity acceleration G _B is represented by each of the acceleration component with respect to the three axes, it equal to the equation (1) below.

In G _B , the superscript T indicates transpose of the vector, and G _B indicates a size of 3 x 1 vector. This acceleration component will vary according to a defined coordinate system, and the present disclosure describes the conversion (conversion to G _L in G _B) of a to distinguish the acceleration of gravity G _B as one of the acceleration components on the gravitational acceleration axis Coordinates focus decide to do

A transformation matrix may be used to transform information expressed between coordinate systems. Coordinate transformation is performed using the transformation matrix C as in Equation 2 below.

According to FIG. 6, and in the equation (2) Coordinates of G _B is the body coordinate system, the coordinate system of G _L it is the local coordinate system. There may be various reasons for converting the body coordinate system to the local coordinate system. For one reason of coordinate transformation, the moving body coordinate system is often information obtained through a sensing unit on a moving body device such as a smartphone, and the local coordinate system is the difference between the navigation frame and the yaw angle, which is often used in navigation technology. Except for , it is the same concept, and it is necessary to perform coordinate transformation for the purpose of easily understanding and calculating the user's location.

For a body coordinate system, may be included in both the components of the gravitational acceleration G _B with respect to the X axis, Y axis and Z axis. Therefore, there is an acceleration generated when the user moves, but conceptually, the gravitational acceleration component and the component other than the gravitational acceleration cannot be clearly distinguished conceptually in the body coordinate system. On the other hand, in the case of the local coordinate system, by separating only the gravitational acceleration component G _B gravitational acceleration appearing on the vertical axis (Z _L) of the local coordinate system on the horizontal plane perpendicular to the gravitational acceleration, that is, on the X _L-axis, Y _L axis users Since the acceleration component in the moving direction can be easily expressed, there is an advantageous effect of accurately measuring the misaligned angle of the device with respect to the user's moving direction.

Therefore, the coordinate system of G _B in the body coordinate system, by representing the coordinate system of G _L to the local coordinate system as, Equation (3) in the case of the coordinate transformation.

When the coordinate transformation matrix is specified with respect to Equation 3 above, Equation 4 is shown below.

(φ is the roll angle, θ is the pitch angle, ψ is the yaw angle)

The roll and pitch angles in the body coordinate system shown in FIG. 6 may be derived by Equations 5 to 6 below. In Equation 4, since there is no z-axis rotation transformation between the body coordinate system and the local coordinate system, the yaw angle becomes 0.

When the coordinates are converted according to Equation 4, accurate roll, pitch, and yaw angle calculations are possible when the calculated acceleration component includes only the gravitational axis acceleration component. Since the component and the horizontal axis acceleration component are included, there is a problem that an error may occur in the coordinate transformation. Accordingly, there is a need for a method capable of accurately separating the gravitational axis acceleration component from the acceleration information sensed by the device.

7 illustrates a process of determining a rotation angle of a device according to an embodiment of the present invention.

In operation S710, the device may receive acceleration information on the movement of the device from the acceleration sensor. Acceleration information sensed by an acceleration sensor built into the device may be acquired, and the acceleration information may include a gravitational axis acceleration component, a vertical axis acceleration component, and a horizontal axis acceleration component.

In the present disclosure, the vertical acceleration component refers to an acceleration component caused by a change in height that occurs while a user of the device moves, and the horizontal axis acceleration component refers to an acceleration component with respect to the moving direction of the user (or device). it means.

In step S720, the device may convert the coordinates of the device by using only the gravitational axis acceleration information in the acceleration information. The acceleration information includes vertical and horizontal acceleration components in addition to the gravitational axis acceleration component, but only the gravitational acceleration component can be used separately. When the acceleration component received on the device is expressed on the moving body coordinate system, it can be expressed in the moving body coordinate system through coordinate transformation.

In operation S730, the device may remove sensor noise information by filtering the acceleration information expressed on the local coordinate system. The vertical and horizontal acceleration components of the acceleration components sensed by the accelerometer according to the user's movement can change with a certain period. have. The device may calculate the frequency by deriving it from the frequency of the magnitude of the acceleration. When the user moves forward, as a characteristic due to dynamics, the frequency of the vertical axis and the horizontal axis has the same frequency as the frequency of the accelerometer size.

In this case, the horizontal acceleration information may be acceleration information for a moving direction of the device, and the vertical acceleration information may be acceleration information between the device and a horizontal plane.

Hereinafter, a method of determining the misaligned angle (rotation angle) of the device by analyzing the change in the acceleration component along each axis on the device according to the user's moving direction will be described.

8 illustrates a change in vertical axis height according to a user's moving direction according to an embodiment of the present invention.

As shown in FIG. 8 , the user may move in a certain direction (horizontal axis direction). Since the user walks using the two legs 7 and 8, the center of gravity changes according to the crossing steps of the two legs. As the center of the user changes, the center of the device held by the user or moved together with the user also changes. Accordingly, the acceleration component of the device changes according to the movement of the user.

Therefore, in order to analyze the change of the acceleration component according to the user's moving direction, each time point of the user's walking pattern will be described. First, in order to analyze the vertical acceleration component, a change in the height of the user's waist will be described.

At time point 1, the user steps on the user's right leg 8 with respect to the direction of travel. When the user steps forward with the right leg 8 , the left leg 7 is relatively positioned behind and the height of the user's waist 6 is the lowest due to the difference in distance between the left leg 7 and the right leg 8 . have a height

At time point 2, when the user steps forward using the right leg 8 in the direction of travel, a time point at which the left leg 7 and the right leg 8 cross each other occurs. In this case, since the left leg 7 and the right leg 8 intersect, the height of the user's waist 6 gradually increases compared to the previous time point 1 and has a maximum value.

At time point 3, after the left leg 7 and the right leg 8 cross each other, the left leg 7 comes forward and the right leg 8 is positioned relatively behind. In this case, the height of the user's waist 6 is lowered again due to the difference in distance between the left leg 7 and the right leg 8, as in time point 1 above, to have a minimum value.

At time point 4, the time point at which the left leg 7 and the right leg 8 cross each other occurs. In this case, since the left leg 7 and the right leg 8 intersect, the height of the user's waist 6 is gradually increased compared to the previous time point 3 and becomes the same as the height at the time point 2 .

Time point 5 has the same type as time point 1 described above. The user steps on the user's right leg 8 with respect to the direction of travel. When the user steps forward with the right leg 8 , the left leg 7 is relatively positioned behind and the height of the user's waist 6 is the lowest due to the difference in distance between the left leg 7 and the right leg 8 . have a height

Analyzing the change in the height of the user at time point 1, time point 3, and time point 5, it can be seen that the pattern of time point 1, time point 3, and time point 5 is the same, and from time point 1 to time point 4 is one cycle. It can be seen that the waist height changes in a pattern similar to a sine wave or a cosine wave. At this time, the height of the user's waist has a maximum value at time points 2 and 4 when the two legs cross each other, and has a minimum value at time points 1, 3, and 5 when the two feet have the largest distance difference from each other.

9 illustrates a change in vertical axis speed according to a user's moving direction according to an embodiment of the present invention.

Since the speed can be derived by differentiating the distance with respect to time, the vertical axis speed component can be derived by differentiating the change in the height of the user's waist in FIG. 8 with respect to time. Therefore, in FIG. 9 , a change in the vertical axis speed according to the passage of time is to be analyzed.

First, in the section between time point 1 and time point 2, the height of the user's waist increases and the vertical axis speed has a positive value based on 0 (assuming that the user moves in a stationary state). The maximum value of the velocity is at the midpoint between the two time points, and after the maximum value is reached, the vertical axis velocity decreases.

In the section between time points 2 and 3, the height of the user's waist corresponds to a section in which the height of the user's waist decreases from the maximum to the minimum, and the vertical axis velocity has a negative value. The minimum value of the velocity is obtained at the intermediate time point between the two time points, and after the minimum value is reached, the vertical axis speed increases again.

In the section between time points 3 and 4, the height of the user's waist increases again from the minimum to the maximum, and the vertical axis velocity has a positive value. The maximum value of the velocity is at the midpoint between the two time points, and after the maximum value is reached, the vertical axis velocity decreases again.

In the section between time points 4 and 5, the height of the user's waist is a section in which the height of the user's waist decreases from the maximum to the minimum, and the vertical axis velocity has a negative value. The minimum value of the velocity is obtained at the intermediate time point between the two time points, and after the minimum value is reached, the vertical axis speed increases again.

The change in vertical velocity in the section between time point 1 and time point 5 also has a pattern similar to that of a sine wave or a cosine wave with the passage of time. The vertical axis velocity has a maximum value at the midpoint of time points 1 and 2, and a midpoint between time points 3 and 4, and a minimum value at the midpoint between time points 2 and 3 and intermediate time points 4 and 5.

10 illustrates a change in vertical acceleration according to a user's moving direction according to an embodiment of the present invention.

Since the speed can be derived by differentiating the distance with respect to time, and the acceleration can be derived by differentiating the speed with respect to time again, the vertical axis acceleration component can be derived by differentiating the user-centered vertical speed change with respect to time in FIG. can Accordingly, in FIG. 10 , a change in the vertical acceleration according to the passage of time is to be analyzed.

First, in the section between time point 1 and time point 2, the value of acceleration, which is the differential value of speed with respect to time, decreases from positive to negative. In the section between time point 2 and time point 3, the value of acceleration, which is the derivative of velocity with respect to time, increases from negative to positive again.

Since the section between time points 3 and 4 has the same pattern as the section between time points 1 and 2 described above, it will be referred to.

The section between time points 4 and 5 has the same pattern as the section between time points 2 and 3 described above, so it will be referred to.

The change in vertical acceleration in the section between time point 1 and time point 5 also has a pattern similar to that of a sine wave or a cosine wave with the passage of time. The vertical acceleration has maximum values at time points 1, 3, and 5, and has a minimum value at time points 2 and 4. In addition, it can be seen that the user's waist height change and the vertical axis acceleration change have a phase difference of 180 degrees from each other. The user's height, speed, and acceleration with respect to the vertical axis may be shown in a form similar to a sine wave or a cosine wave, repeating the section from time point 1 to time point 3 above as one cycle, and repeating maximum and minimum values.

Similar to the previous review of the vertical acceleration change in FIGS. 8 to 10 , the horizontal acceleration change with respect to the user's moving direction will be reviewed.

11 illustrates a change in horizontal acceleration according to a user's moving direction according to an embodiment of the present invention. Since the horizontal axis is the same as the user's moving direction, the analysis of distance and speed will be omitted.

In the section between time point 1 and time point 2, the horizontal acceleration component has a negative value. That is, the speed in the moving direction of the user gradually decreases. Since this is a section in which the left leg 7 is pulled until the two legs cross by the right leg 8 stepping forward, the speed is reduced.

In the section between time points 2 and 3, the weight of the user is added to the right leg 8 to step on and propelling in the moving direction, so the acceleration value has a positive value. This is the point at which the user's center of gravity starts to be located in front of the right leg, and the reason for the positive acceleration is that when the user's center of gravity is greater in front than in the back, when the leg on the floor pushes the body with the same force. It can be interpreted in terms of being able to generate horizontal thrust. In the previous section (between time points 1 and 2), the user's center of gravity is behind the right leg (8) that touches the floor, and the left leg (7) is in the air, so if propulsion occurs in this section In order to lift the user's center of gravity in the upward direction against gravity, the leg that touches the floor must apply force not only on the horizontal axis but also on the vertical axis. That is, we need another force that opposes gravity. Therefore, in this section, it is advantageous to use the inertia of the body to move forward rather than moving forward with force. On the other hand, in the section between time points 2 and 3, the direction of the vertical axis of the user's center of gravity is the same as the direction of gravity. Therefore, in this section, since the vertical acceleration generated when the center of gravity freely falls is added to the horizontal acceleration generated when the user proceeds to the horizontal axis, a greater movement effect can be seen compared to the previous section with less force than the previous section. .

Since the section between time points 3 and 4 shows the same shape as the section between time points 1 and 2 described above, and the section between time points 4 and 5 shows the same shape as the section between time points 2 and 3, detailed description is omitted.

It can be seen that the horizontal acceleration component also has a constant period with respect to time like the vertical acceleration component, and can be expressed in a form similar to a sine wave or a cosine wave having repeated maximum and minimum values. From this analytical point of view, the horizontal axis acceleration and the vertical axis acceleration have a form in which the phase is 90 degrees orthogonal to each other.

It should be noted in FIG. 11 that the maximum value of the positive horizontal acceleration component is greater than the absolute value of the minimum value of the negative horizontal acceleration component, which means that the user accelerates when moving forward.

12 shows a case in which the user moves at a constant speed (26) and a case in which the user decelerates (32), respectively.

In the case of constant velocity movement (26), the reason that the magnitude of the absolute value of acceleration in the sections of time points 2 and 3 is the same as the magnitude of the absolute value of acceleration in sections of time 1 and 2 is that the acceleration in the user's moving direction is an acceleration opposite to the moving direction. because it is similar to That is, it is a case of constant velocity motion. Since the section between time points 3 and 5 has the same shape as the section between time points 1 and 3, a detailed description thereof will be omitted.

In the case of deceleration (32), the reason that the magnitude of the absolute value of acceleration in the sections of time 1 and 2 is greater than the magnitude of the absolute value of acceleration in sections 2 and 3 is that the user decelerates and the acceleration in the opposite direction to the moving direction is greater than the acceleration of Since the section between time points 3 and 5 has the same shape as the section between time points 1 and 3, a detailed description thereof will be omitted.

Previously, the vertical axis acceleration component and the horizontal axis acceleration component were analyzed in FIGS. 8 to 12 . Since the accelerometer of the device senses an acceleration component that includes a gravitational axis acceleration component in addition to the vertical acceleration component and the horizontal axis acceleration component, it is necessary to clearly distinguish the gravitational axis acceleration and the vertical/horizontal axis acceleration components.

In order to analyze the rotation angle of the device, that is, roll, pitch, and yaw angle, it is necessary to use the acceleration of the gravitational axis in the local coordinate system as in Equations 4 to 6. Therefore, accurate roll and pitch angle calculations are possible only when the gravitational axis acceleration component is separated from the acceleration information sensed by the accelerometer of the device.

In the present disclosure, a method of separating only the gravitational axis acceleration component from the acceleration component sensed by the accelerometer of the device will be mainly described. It will be reviewed with reference to FIG. 11 above.

In FIG. 11 , the vertical acceleration component and the horizontal acceleration component have a phase difference of 90 degrees from each other and may be represented by a sine wave graph indicating the maximum and minimum values. At this time, it can be seen that when the vertical acceleration component is the maximum value, the horizontal acceleration component has a value of 0. That is, in the figure, the vertical acceleration component at time point 1 and time point 3 represents the maximum value. Since the interval between the respective viewpoints is constant, it can be derived that the vertical acceleration component has a minimum value at the time point 2 that is the center of the viewpoints 1 and 3.

The control unit of the device may obtain the gravitational acceleration component through a method of merging the vertical acceleration component and the horizontal acceleration component at time 1 and time 2 . That is, the horizontal acceleration has a value of 0 at time point 1 when the vertical acceleration is the maximum value, and the horizontal axis acceleration has a value of 0 even at time point 2 when the vertical axis acceleration is the minimum value. The acceleration component also has a value of 0. Therefore, only the gravitational axis acceleration component can be separated from the average value of these two time points. The roll angle and pitch angle of the device can be derived using the gravitational axis acceleration component.

The point in time when the vertical acceleration component has the maximum value is the same as the point in time when the magnitude of the acceleration sensed by the accelerometer of the device has the maximum value. Hereinafter, we will prove that the time point at which the vertical acceleration component expressed in the local coordinate system has the maximum value and the point in time when the acceleration magnitude expressed in the local coordinate system has the maximum value are the same.

When the magnitude of the acceleration with respect to time t is f(t), the square F(t) of the magnitude of the acceleration can be expressed as Equation 7 below in the local coordinate system.

When a ^{value of AccX 2} + AccY ² + AccZ ² is substituted for Equation 7 above, it can be expressed as Equation 8.

로, 수평축 가속도(12)는

로 표현된다. 따라서 도 13에서 좌표계 A에서 표현된 가속도의 크기를 제곱하면 하기 수학식 8로 표현된다.13 shows the coordinate system (A) and the local coordinate system (L) used in the drawings of FIGS. 8 to 12 . Taking Fig. 11 as an example, with respect to the time section from the user's initial position (1) to the user's final position (5), when the user's initial position is defined as the origin, the vertical acceleration 11 is

, the horizontal acceleration (12) is

is expressed as Therefore, squaring the magnitude of the acceleration expressed in the coordinate system A in FIG. 13 is expressed by Equation 8 below.

In Equation (8), f is the user's walking frequency, π is the circumference, t is time, and g is the acceleration of gravity, and is a value greater than 0. The a value indicates the maximum acceleration value for the user's moving direction and is greater than 0, and the b value indicates the maximum acceleration value on the vertical axis excluding the gravitational acceleration when the user moves.

When Equation 8 is expressed in the local coordinate system, only the sign of the vertical acceleration component is changed, and it can be expressed as Equation 9 below. It should be noted here that in both coordinate systems shown in FIG. 13, the magnitudes of the accelerations are the same, but the vertical axis components have opposite signs.

In the present disclosure, acceleration information is obtained and calculated based on the moving direction of the user (or device), but the present invention is not limited thereto, and acceleration information may be calculated based on the direction opposite to the user's moving direction. The angle of rotation can be determined by appropriately converting a positive or negative value for the direction of acceleration. In the present disclosure, with respect to the method of determining the rotation angle with respect to the direction opposite to the traveling direction, it is necessary to note that it is possible to appropriately modify acceleration information and equations in the method for determining the rotation angle with respect to the traveling direction.

14 is a Z-axis acceleration in which the X-axis traveling direction acceleration and gravitational acceleration are removed from the local coordinate system when the user matches the body coordinate system, which is the posture of the device, to the local coordinate system, and matches the user's moving direction with the X axis of the local coordinate system and walks. is an experimental graph.

As shown in FIG. 14 , although some sensor noise is included, the two accelerations are orthogonal to each other. The reason that the vertical acceleration of FIG. 11 is symmetric with respect to the time axis is because, as described in FIG. 13 , the vertical axis direction of the local coordinate system is opposite to the vertical axis direction of the drawing. 14 , the magnitude of the horizontal acceleration means the maximum value of the absolute value of the acceleration when the user accelerates, accelerates, or decelerates. An important point in FIG. 14 is that the magnitude of the vertical acceleration is greater than the magnitude of the horizontal acceleration. This has been confirmed through various experimental results and is used as an important characteristic when proving the process proposed in the present invention through the following equation.

Differentiating Equation 9 with respect to time in order to obtain the maximum and minimum values of F(t) in Equation 9 can be expressed as Equation 10 below.

또는

인 두 가지 경우에 해당되어야 한다.To find the local maxima and minima, if the above derivative becomes 0,

or

It should be true in two cases.

가 되기 위한 경우를 고려할 때,

값이 0인 경우, 양의 값인 경우, 음의 값인 경우로 나누어 생각할 수 있다.first,

Considering the case to become

When the value is 0, it can be divided into a positive value and a negative value.

값이 0인 경우(

)에는 bg값이 0보다 크므로

이고 상기의 조건을 만족하지 못한다.i)

If the value is 0 (

) has a bg value greater than 0.

and does not satisfy the above conditions.

값이 양인 경우(

)를 고려하면,ii)

If the value is positive (

), taking into account

이 되어야 하므로, 하기의 수학식 11과 같이 나타낼 수 있다.

Since it should be, it can be expressed as in Equation 11 below.

If Equation 11 is arranged for a, it can be expressed as Equation 12 below.

이면, if,

this side,

이 될 수 있다.

의 최소값은 분모가 최대가 되는 t = 0일 때가 해당되며,

이 된다.

this can be

The minimum value of is when t = 0 where the denominator is maximum,

becomes this

이 성립하게 된다. 즉,

이 되므로, 상기의

조건을 만족하지 못한다.Through many experimental results comparing the maximum acceleration on the horizontal axis and the maximum acceleration on the vertical axis, since the maximum horizontal acceleration value a is smaller than the vertical maximum acceleration value b,

this will come true in other words,

As it becomes, the above

condition is not satisfied

값이 0보다 작은 경우(

)를 고려하면, iii)

If the value is less than 0 (

), taking into account

가 되기 위해서는 하기의 수학식 13을 만족하여야 한다.From the results of several experiments, if a < b

In order to become , the following Equation 13 must be satisfied.

을 만족할 수 있으므로, 극값을 가질 수 있다.Equation 13 is based on the values of a and b

can be satisfied, so it can have an extreme value.

및 수학식 13이며, 하기의 세가지 경우로 나타낼 수 있다.In conclusion, the condition under which the magnitude of the acceleration can have an extreme value, that is, a maximum/minimum value, is

and Equation 13, and can be represented by the following three cases.

A value of t that satisfies 1), 2) or 3) has an extreme value of F(t). In order to mutually compare the magnitudes of the three extreme values, the magnitudes of the extreme values in 1) and 3) are compared as shown in Equation 14 below.

라고 하면,In Equation 14 above

If you say

이다. 따라서, G(X)는 중근을 가지며, X는 하기의 수학식 15와 같다.Here, for any a and b, the discriminant of G(X)

am. Accordingly, G(X) has a middle root, and X is expressed by Equation 15 below.

Therefore, at X = bg, G(X) has a minimum value of 0 as follows.

In conclusion, the magnitudes of the three extreme values are compared as follows.

The magnitude comparison above refers to a magnitude comparison of the acceleration F(t) for cases 1), 2), and 3) of Equation 13.

경우의 극값이

경우의 극값보다 크고,

경우의 극값이

경우의 극값보다 크거나 같다는 의미이다.in other words,

the extreme value of the case

greater than the extreme value of the case,

the extreme value of the case

It means that it is greater than or equal to the extreme value of the case.

At this time, (a+b) ² means the maximum and maximum value of F(t), and it can be proved as follows.

If F(t) is differentiated twice with respect to time t in order to obtain the maximum and minimum values of F(t), the following Equation 16 is obtained.

이며, 이때의 F(t)에 대하여 두 번 미분하면 하기와 같다.thus,

, and differentiating it twice with respect to F(t) at this time is as follows.

Therefore, under the above conditions, F(t) has a maximum value.

의 의미는 도 10에서는 수직축 방향이 반대로 정의가 되었기 때문에

인 지점을 의미한다. 즉 사용자의 두 다리가 서로 교차하는 시점(사용자의 허리 높이의 최대값이 발생하는 시점)에서 수직축 가속도의 크기가 최대값을 갖고 동시에 3축 가속도의 크기(가속도 벡터의 2 norm값) F(t)가 최대값을 가짐을 증명한 것이다.Here, F(t) is a value obtained by squaring the magnitude of the acceleration in Equation 7, and the point at which the maximum value of the square of the magnitude of the acceleration is generated is the same as the point at which the maximum value of the magnitude of the acceleration is generated. In the local coordinate system, the vertical axis is defined in the direction from the user's head to the legs and is opposite to the vertical axis direction defined in FIG. 10 . Therefore, the formulas proven so far are all proven in the local coordinate system, and in the local coordinate system,

The meaning of is because the vertical axis direction is defined oppositely in FIG.

means a point. That is, at the point where the user's two legs cross each other (the point at which the maximum value of the user's waist height occurs), the magnitude of the vertical acceleration has the maximum value and at the same time the magnitude of the 3-axis acceleration (the 2 norm value of the acceleration vector) F(t) ) proves that it has a maximum value.

항으로 표현되는 수직축 가속도의 크기의 최대값과 동일함을 도출할 수 있다. 다만, 한가지 유의할 점은 수직축 가속도 크기의 최소값은 가속도 크기의 최소값과 다를 수 있다는 점이다. 따라서, 가속도 크기의 최소값을 가지는 시점은, 최소값과 인접하고 가속도 크기가 최대인 두 최대값의 시점을 평균하여 획득할 수 있다.From the above proof process, the maximum value of the magnitude of the acceleration sensed by the accelerometer of the device is the gravitational acceleration and

It can be derived that is the same as the maximum value of the magnitude of the vertical acceleration expressed in terms. However, it should be noted that the minimum value of the vertical acceleration magnitude may be different from the minimum value of the acceleration magnitude. Accordingly, the viewpoint having the minimum value of the acceleration magnitude may be obtained by averaging the viewpoints of the two maximum values adjacent to the minimum value and having the largest acceleration magnitude.

The above-described theoretical proof process can be verified through simulation. The size of the maximum acceleration in the vertical direction from 0.1m / s ² is changed at intervals of 0.1m / s ² to 30m / s ^2, the horizontal axis acceleration was changed to 0.1 times the interval from 0.1 times to 0.9 times of the acceleration. This reflects the experimental result that the maximum value of the vertical acceleration is larger than the maximum value of the horizontal acceleration. In addition, the vertical and horizontal triangular waveforms were simulated at intervals of 0.01 degree for one period.

In the simulation results, the point at which the maximum value of the vertical acceleration occurred and the point at which the maximum value of the magnitude of the acceleration occurred were the same for all cases. However, there existed a case where the time at which the minimum value of the vertical acceleration occurred was different from the time at which the maximum value of the magnitude of the acceleration was generated.

15 is a graph showing that the time point at which the minimum value of the vertical acceleration appears in the local coordinate system may be different from the time point at which the minimum value of the magnitude of the acceleration appears in the simulation.

One important point in the proof is that all these analyzes were performed in the local coordinate system, but the accelerations given by the accelerometer are all expressed in the body coordinate system. Here, the point at which the maximum value of the vertical acceleration is generated in the local coordinate system is the same as the point in time at which the maximum value of the magnitude of the acceleration is generated in the local coordinate system. The reason is that the magnitude of the acceleration sensed by the 3-axis accelerometer is expressed at the same time point at which the maximum or minimum value is expressed whether it is expressed in the local coordinate system, in the body coordinate system, or in any other 3-axis vertical coordinate system. Therefore, only the gravitational acceleration can be extracted from the maximum value of the 3-axis acceleration value of the accelerometer sensed in the body coordinate system (2-norm value of the 3 x 1 acceleration vector), and the roll and pitch angles of the device can be obtained.

Since the time point information having the maximum value and the time point information having the minimum value can be obtained, the control unit of the device averages the acceleration values for each axis with respect to these two time points, the vertical axis acceleration component and the horizontal axis acceleration component are removed from the acceleration components, and gravity Acceleration information in which only an axial acceleration component exists can be acquired.

In conclusion, by using the dynamic characteristics according to the movement of the user, only the gravitational acceleration component is separated from the acceleration components for each axis of the accelerometer of the device, and the correct roll angle and pitch angle using Equations 5 and 6 can be calculated. By substituting the calculated roll angle and pitch angle into Equation 4, the coordinate transformation matrix can be derived. By obtaining the horizontal acceleration expressed on the Y-axis, it is possible to obtain a misaligned angle on the horizontal plane of the device with respect to the user's moving direction.

16 is a diagram illustrating a method of filtering an acceleration component in order to minimize an error due to noise when obtaining a maximum value of an acceleration magnitude in a moving body coordinate system according to an embodiment of the present invention.

There is a problem in that the rotation angle information of the device measured based on the gravitational axis acceleration component in the accelerometer of the device may include noise. Accordingly, it is possible to reduce the error in the rotation angle of the device to be obtained by filtering the measured acceleration information.

It has been described above that the acceleration changes with a certain period according to time according to the movement of the user. Accordingly, the vertical acceleration and the horizontal acceleration may have respective frequencies. For convenience of description, it will be referred to as “walking frequency”. When the user moves, the user can cross the left leg and the right leg with a constant cycle, but it is not always constant. Therefore, by obtaining the average cycle for a certain section, the user's walking frequency is adaptively can be estimated.

In FIG. 16 , when the average value of acceleration is obtained for a time point 29 having a maximum value and a time point 30 having a minimum value on the graph representing the vertical acceleration magnitude, the gravitational acceleration is extracted, and an error may occur due to sensor noise. Therefore, the accelerations of the two viewpoints can be averaged through the time points before and after by the section T based on the time points 29 and 30, respectively. When the average acceleration value obtained from the body coordinate system is obtained through this filtering process, the vertical acceleration average is canceled by the sum of the maximum and minimum values, and since the horizontal acceleration is 0, only the gravitational acceleration with reduced error due to noise remains. If the magnitudes of the positive and negative accelerations on the horizontal axis are similar to each other, the gravitational acceleration can be removed by increasing T to half a period and averaging the entire period. If the magnitudes of the positive and negative accelerations on the horizontal axis are similar to each other, it can be extended to a method of removing the gravitational acceleration by averaging the N periods by increasing T by N/2 periods for a natural number N.

17 is a diagram showing the magnitude of the 3-axis acceleration, the magnitude when the gravitational acceleration is removed, and the true value of the gravitational acceleration in the body coordinate system.

As for the magnitude of the 3-axis acceleration, vertical and horizontal acceleration occur when the user proceeds, and a large error similar to a sine wave occurs based on the true value of the gravitational acceleration. However, the proposed method shows that the vertical acceleration and the horizontal acceleration for the user's moving direction are effectively removed to extract a value almost similar to the true value of the gravitational acceleration. This is consistent with the theoretical conclusion that the method proposed in the present invention provides an optimal solution through mathematical proof. If the acceleration without removing the user's vertical and horizontal acceleration from the body coordinate system is used to calculate the roll and pitch, it can be expected that a large posture error will occur. This posture error makes the coordinate conversion error large and eventually causes an X and Y-axis acceleration error in the local coordinate system, which means that the device's angular error with respect to the user's moving direction is generated.

Through the above process, by extracting only the gravitational acceleration among the acceleration components in the body coordinate system to obtain the roll and pitch angle of the device, and substituting it into the coordinate transformation matrix of Equation 4, the X, Y acceleration values in the local coordinate system can be obtained through the coordinate transformation. have. To obtain the X and Y accelerations in the local coordinate system, in Equation 4, the right side of the coordinate transformation 3 x 3 matrix C is calculated by multiplying ^{the 3 x 1 acceleration vector [X-axis acceleration, Y-axis acceleration, Z-axis acceleration] T of the body coordinate system.} The resulting 3 x 1 acceleration vector means the acceleration in the local coordinate system. Therefore, among the 3 x 1 acceleration vectors in the local coordinate system, the first two terms are the X and Y acceleration values in the local coordinate system.

Although it is possible to obtain a misaligned angle of the horizontal axis device with respect to the user's moving direction from the X and Y acceleration values in the local coordinate system, an angular error may occur due to sensor noise. The present disclosure proposes to use a bandpass filtering technique as a method for removing such noise.

18 is a diagram illustrating a method of removing noise from X-axis and Y-axis acceleration expressed in a local coordinate system.

In FIG. 18, viewpoints 50, 51, and 52 show a process of adaptively estimating the frequency of the acceleration magnitude obtained from the body coordinate system according to time. The frequency can be obtained from one step cycle, or the frequency can be obtained from the N step cycle. This frequency is used as the center frequency 53 of the bandpass filter, and it can be designed to vary the center frequency according to the user's step frequency on the time axis. The bandwidth 54 of the bandpass filter is also an important parameter, and the bandwidth can be determined through tuning. By bandpass filtering the X and Y axis acceleration expressed in the local coordinate system, it is possible to obtain an acceleration with reduced sensor noise.

19 shows a misaligned angle of a device according to an embodiment of the present invention.

The device may not always be positioned in the same direction with respect to the user's movement (progression) direction, but may be positioned with an erroneous (rotational) angle. As shown in FIG. 19 , the user's moving direction may be positioned at a constant angle on the X axis and the Y axis on the local coordinate system. Assuming that the acceleration value on the X-axis to which bandpass filtering is applied in the local coordinate system is Ax, and the acceleration value on the Y-axis to which the bandpass filtering is applied in the local coordinate system is Ay, the user on the horizontal plane of the X-axis and Y-axis of the local coordinate system The misaligned angle of the device with respect to the traveling direction of , may be expressed as in Equation 17 below.

By measuring the misaligned angle of the device as in Equation 17 above, it can be applied to various applications within the device.

20 is a graph obtained experimentally from small errors to large errors of the device's misaligned angular estimation error with respect to the user's moving direction on a horizontal plane perpendicular to the gravitational acceleration made up of the X and Y axes of the local coordinate system.

Figure 20 (a) shows the result of estimating the wrong angle of the mobile phone with respect to the user's moving direction by extracting the gravitational acceleration by the method proposed in the present invention, and the estimation using the acceleration including the vertical and horizontal accelerations in addition to the gravitational acceleration. the results are shown. 20 (b) is a graph showing an enlarged measurement error range of the graph (a). As shown in (a) of FIG. 20 , when acceleration components other than gravitational acceleration are not properly removed, a posture error of the device is generated, and each component of the horizontal acceleration is greatly distorted through incorrect coordinate transformation, resulting in a very large estimation error. Each estimation error in the drawing is a result derived using data obtained through experiments, not simulations.

As shown in FIG. 20 , a total of 26 tests were performed for 8 angular errors from 0° to 315° at intervals of 45°. 1354 steps occurred from all tests, and each error for each step is shown in order of magnitude. The average of the angle error was 1.2 degrees and the standard deviation was 9.3 degrees, and this result was verified through an actual test, the theoretical method of obtaining the misaligned angle of the device with respect to the user's moving direction using only the accelerometer in the present invention can be considered a result.

21 is a block diagram illustrating the configuration of a device according to an embodiment of the present invention.

The device 100 according to an embodiment of the present invention may include a control unit 110 , a sensing unit 120 , an input/output unit 130 , and a storage unit 140 . Additionally, it may further include a display unit (not shown) for displaying device information and a communication unit (not shown) for communicating with other devices.

The sensing unit 120 may acquire various external information outside the device 100 . It may include an accelerometer 125 that acquires the acceleration of the device, and may measure various environmental information such as temperature, humidity, location, and pressure.

The input/output unit 130 may include an input unit for receiving information and commands received from the outside of the device 100 and an output unit for transmitting information and commands processed inside the device 100 to the outside.

The storage unit 140 may store information processed inside the device 100 , and may store information received outside the device 100 .

The device 100 according to an embodiment of the present invention may be a smart phone capable of executing Internet access or various execution programs by embedding an operating system. Smartphones are equipped with operating systems and communication functions in digital portable devices so that various contents can be used in a convenient user environment (UI/UX). In addition, devices such as a multimedia player and a personal computer may correspond to the terminal of the present invention.

The control unit 110 according to an embodiment of the present invention may process the processes of Equations 1 to 17 above with respect to acceleration information sensed by the device. Various information input to the device may be performed according to a preset processing process.

The controller 110 may transform the coordinates of the acquired acceleration information, filter the acceleration information on the transformed coordinates, and determine the rotation angle with respect to the moving direction of the device using the gravitational axis acceleration information on the transformed coordinates. . In this case, the gravitational axis acceleration information may be information derived from the acceleration information using information about a time point in which the vertical and horizontal acceleration components with respect to the moving direction of the device are maximum.

The controller 110 may convert the obtained acceleration information from coordinates on a coordinate system based on the device to coordinates on a coordinate system based on a gravity direction.

The controller 110 may filter the acceleration information after converting it into a frequency domain. In this case, the filtering may be bandpass filtering. More specifically, the controller 110 may perform bandpass filtering from the frequency domain-converted acceleration information based on a center frequency determined using the dynamic characteristics of the device 100 .

The control unit 110 encodes or decodes data input through the input/output unit 130 . The controller 110 provides a user interface based on an operating system (OS) of the device 100 . Such a user interface may reflect a user's usage mode.

The display unit (not shown) may display data processed by the device in a user interface environment. In addition, when a terminal user's operation command is requested, information for guiding to input various operation commands may be displayed on the screen.

A communication unit (not shown) transmits and receives data and control commands from other devices. As the communication unit 150 , a known communication module such as an infrared communication module, a radio communication module, or an optical communication module may be used. For example, an infrared communication module that satisfies an infrared communication standard IrDA (Infrared Data Association) protocol may be used as the communication unit. As another example, a communication module using a 2.4 GHz frequency or a communication module using Bluetooth may be used as the communication unit.

The sensing unit 120 according to an embodiment of the present invention may sense measurable elements from inside or outside the device. The sensing unit 120 according to an embodiment of the present invention may include an accelerometer 125 , and the accelerometer 125 may acquire the motion of the device in a 3D space as acceleration information.

The input/output unit 130 according to an embodiment of the present invention may serve to receive a command from the outside of the device and output processed information. For example, in the case of a device such as a smartphone, a key or touch input may be received, and the processed information may be output in a way that is displayed on the screen or may be output as sound.

The input unit of the input/output unit 130 is an interface for receiving data such as content displayed on the display unit, and is a Universal Serial Bus (USB), Parallel Advanced Technology Attachment (PATA), or Serial Advanced Technology Attachment (SATA). ), flash media, Ethernet, Wi-Fi, and Bluetooth)). In some cases, the device 100 may be provided with an information storage device (not shown) such as an optical disk drive or a hard disk to receive data through it.

Also, the device 100 according to an embodiment of the present invention may include a camera unit including an optical image sensor. An image received through the camera module may be processed as one data.

The input unit may further include a sensor unit such as temperature and humidity in addition to the camera unit. The temperature sensor unit for measuring the internal temperature of the device may also include a temperature sensor unit for measuring the external temperature of the device, and the humidity measuring unit for measuring humidity may also correspond to the input unit.

Also, the input unit may be a touch screen in which the touch panel unit and the image panel unit have a layered structure. The touch panel unit may be, for example, a capacitive touch panel, a resistive touch panel, or an infrared touch panel. The image panel unit may be, for example, a liquid crystal panel, an organic light emitting panel, or the like. Since such a touch screen panel is well known, a detailed description of the panel structure will be omitted. The image panel unit may display a graphic of a user interface.

The storage unit 140 according to an embodiment of the present invention may store information processed by the control unit 110 . In addition, information sensed by the sensing unit 120 may be stored, and information received from other devices may be stored. Such information may be classified and stored as a database (DB) in the storage unit 140 .

Meanwhile, the above-described embodiments of the present invention can be written as a program that can be executed on a computer, and can be implemented in a general-purpose digital computer that operates the program using a computer-readable recording medium. The computer-readable recording medium includes a storage medium such as a magnetic storage medium (eg, a ROM, a floppy disk, a hard disk, etc.) and an optically readable medium (eg, a CD-ROM, a DVD, etc.).

So far, the present invention has been looked at with respect to preferred embodiments thereof. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments are to be considered in an illustrative rather than a restrictive sense. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

100: device

Claims (32)

In the method of determining the rotation angle with respect to the traveling direction of the device,
obtaining acceleration information on the movement of the device from an acceleration sensor of the device;
transforming the coordinates of the acquired acceleration information of the device;
filtering the acceleration information on the transformed coordinates of the device; and
determining a rotation angle with respect to the moving direction of the device by using the information on the acceleration of the gravitational axis on the converted coordinates;
including,
The gravitational axis acceleration information includes a first average acceleration value in a section before and after a point in time when the vertical acceleration component on the converted coordinates of the device in the acceleration information is maximum, and a first average acceleration value in a section before and after the point in time when the vertical acceleration component is the minimum. 2 A method for determining a rotation angle, which is determined based on an average value of the average acceleration values. According to claim 1,
A point in time at which the vertical acceleration component on the transformed coordinates is maximum is the same as a point in time when the magnitude of the obtained acceleration has a maximum value. According to claim 1,
The minimum value of the vertical acceleration component on the transformed coordinates is,
A method for determining a rotation angle derived using information about a time point having a maximum value of the vertical acceleration component on the transformed coordinates. According to claim 1,
The horizontal acceleration component included in the acceleration information is acceleration information for a moving direction of the device, and the vertical acceleration component is an acceleration component between the device and a horizontal plane. According to claim 1,
The step of transforming the coordinates of the obtained acceleration information comprises:
A method for determining a rotation angle, characterized in that the acceleration information is converted from coordinates on a coordinate system based on a device to coordinates on a coordinate system based on a gravity direction. 6. The method of claim 5,
The coordinate system based on the direction of gravity includes three axes orthogonal to each other,
The first of the three axes is an axis parallel to the direction of gravitational acceleration. 7. The method of claim 6,
A second axis of the three axes is defined by rotating one axis on the device reference coordinate system by a roll angle, and a third axis of the three axes rotates another axis on the device reference coordinate system by a pitch angle A method of determining the angle of rotation, which is defined by According to claim 1,
A method for determining a rotation angle, characterized in that the vertical acceleration component and the horizontal acceleration component on the converted coordinates change with the same periodic value. According to claim 1,
A method for determining a rotation angle, characterized in that the vertical acceleration component and the horizontal acceleration component on the converted coordinates have a phase difference of 90 degrees from each other. 10. The method of claim 9,
The rotation angle determination method, characterized in that the vertical axis acceleration component is 90 degrees faster than the horizontal axis acceleration component. 10. The method of claim 9,
The rotation angle determination method, characterized in that the vertical axis acceleration component is 90 degrees slower than the horizontal axis acceleration component. According to claim 1,
The gravitational axis acceleration information is derived by averaging the magnitudes of vertical and horizontal acceleration components on the transformed coordinates of the device over a plurality of periods. According to claim 1,
The step of filtering the acceleration information includes:
The method further comprising converting the acceleration information into a frequency domain. 14. The method of claim 13,
The step of filtering the acceleration information includes:
A method for determining a rotation angle, characterized in that the acceleration information is bandpass filtered. 15. The method of claim 14,
The bandpass filtering comprises:
In the acceleration information converted to the frequency domain, the method of determining a rotation angle, characterized in that filtering based on a center frequency determined using a dynamic (dynamic) characteristic of the device. 16. The method of claim 15,
The dynamic characteristics of the device are,
A method of determining a rotation angle obtained by a posture in which the device user steps in the traveling direction. A device for determining a rotation angle with respect to a traveling direction, the device comprising:
an acceleration sensing unit to obtain acceleration information on the movement of the device;
a storage unit for storing the obtained acceleration information; and
comprising a control unit;
The control unit is
Transform the coordinates of the obtained acceleration information, filter the acceleration information on the transformed coordinates, and determine a rotation angle with respect to the moving direction of the device using the gravitational axis acceleration information on the transformed coordinates,
The gravitational axis acceleration information includes a first average acceleration value in a section before and after a point in time when the vertical acceleration component on the converted coordinates of the device in the acceleration information is maximum, and a first average acceleration value in a section before and after the point in time when the vertical acceleration component is the minimum. 2 The rotation angle determining device, which is determined based on an average value of the average acceleration values. 18. The method of claim 17,
A point in time at which the vertical acceleration component on the transformed coordinates is maximum is the same as a point in time when the magnitude of the obtained acceleration has a maximum value. 18. The method of claim 17,
The control unit, the minimum value of the vertical acceleration component on the transformed coordinates,
A device for determining a rotation angle, which is derived using information about a time point having a maximum value of the vertical acceleration component on the transformed coordinates. 18. The method of claim 17,
The horizontal acceleration component included in the acceleration information is acceleration information for a traveling direction of the device, and the vertical acceleration component is an acceleration component between the device and a horizontal plane. 18. The method of claim 17,
the control unit
A device for determining a rotation angle, characterized in that the obtained acceleration information is converted from coordinates on a coordinate system based on a device to coordinates on a coordinate system based on a gravity direction. 22. The method of claim 21,
The coordinate system based on the direction of gravity includes three axes orthogonal to each other,
The first of the three axes is an axis parallel to the direction of gravitational acceleration. 23. The method of claim 22,
A second axis of the three axes is defined by rotating one axis on the device reference coordinate system by a roll angle, and a third axis of the three axes rotates another axis on the device reference coordinate system by a pitch angle Defined by, rotation angle determining device. 18. The method of claim 17,
The rotation angle determining device, characterized in that the vertical acceleration component and the horizontal acceleration component on the transformed coordinates change with the same periodic value. 18. The method of claim 17,
A rotation angle determining device, characterized in that the vertical acceleration component and the horizontal acceleration component on the transformed coordinates have a phase difference of 90 degrees from each other. 26. The method of claim 25,
The rotation angle determining device, characterized in that the vertical acceleration component is 90 degrees faster than the horizontal acceleration component. 26. The method of claim 25,
The rotation angle determining device, characterized in that the vertical acceleration component is 90 degrees out of phase with the horizontal acceleration component. 18. The method of claim 17,
The gravitational axis acceleration information is derived by averaging the magnitudes of vertical and horizontal acceleration components with respect to the traveling direction of the device over a plurality of periods, the rotation angle determining device. 18. The method of claim 17,
The control unit is
The rotation angle determination device, characterized in that the filtering is performed after converting the acceleration information into a frequency domain. 30. The method of claim 29,
The control unit is
A device for determining a rotation angle, characterized in that the acceleration information is bandpass filtered. 31. The method of claim 30,
The control unit is
In the acceleration information converted into the frequency domain, bandpass filtering is performed based on a center frequency determined using a dynamic characteristic of the device. 32. The method of claim 31,
The dynamic characteristics of the device are,
A rotation angle determination device, which is obtained by a posture in which the device user steps in the traveling direction.

Images